Continuous Monitoring System for the Wastewaters Having Multiply, Randomly, and Small Effluent Characteristics

(1)

Continuous Monitoring System for the Wastewaters Having Multiply, Randomly, and Small Effluent Characteristics

-Approarch to AnalysisofChemical Oxygen Demand by Complete Flow Process-

* ** . ***

Takashi KORENAGA, Tosio MORIWAKE, and Teruo TAKAHASHI

(Received September 17, 1985)

Synopsis

A simple system was developed for the fully

automatic and continuous measurement of chemical oxygen demand (COD) in wastewater samples based on colorimetry of dichromate. A sample and a solution of sulfuric acid

(1+1) containing 2mM potassium dichromate are continuously pumped with a double-reciprocating micro-pump at each flow rate of 0.3 ml/min. The wastewater sample is filtered at first with a 100-mesh stainless filter and then mixed with the dichromate solution in the mixing joint. The mixture is introduced into a reaction coil made of poly(tetrafluoroethylene) tubing (1 mm i.d., 3 mm o.d., and 20 m length), being placed in an oil bath

(120 °C). After reaction, the mixture passes into a quartz tubular flow-through cell (10 mm path length, 18 Jll volume) in a spectrophotometer, and the absorbance is measured at 445 nm. The COD value of the sample is automatically estimated from the amount of decreased absorbance. The system was successfully applied to COD measurement of some waters, and to continuous monitoring of COD in wastewater of university laboratories. The system was also evaluated by comparing with the flow injection analyzer system previously developed by the authors.

*

Center for Environmental Science and Technology, Okayama University

** Department of Synthetic Chemistry

***Department of Industrial Chemistry

45

(2)

46 Takashi KORENAGA, Toshio MORIWAKE and Teruo TAKAHASHI

Introduction

Environmental Scientists are recently aware of the increasing need for automated methods which are suitable for continuous monitoring of wastewater qualities, such as chemical oxygen demand (COD), total phosphorus, and total nitrogen. The authors previously reported some continuous monitoring systems for COD in wastewaters based on flow injection analysis (FIA) with permanganate~1)-(3) dichromate~2)-(4) and cerium(IV)~3),(s)Also, Ishii et. ale reported a continuous monitoring system for COD in waters based on flow amperometry of permanganate~6)-(8)

In the automatic batch analyzer for COD!9) each discrete sample is assigned a container or sample cup, in which it is held during all the steps necessary to perform the analysis. Individual samples are transported through a number of stations where manual operations (such as pipetting, dilution, reagent addition, mixing, heating, reaction, extraction, etc.) and electrochemical measurement, including titration, are mechanized. This simulates a manual procedure. The advantage of the batch approach is that analytical results correlate well with those obtained by the already optimized method as standard. Its disadvantage is that the instrument required is mechanically very complex with many moving parts and each measurement usually requires a large amount of reagents and about an hour.

The present paper therefore describes a simple continuous monitoring system for COD in wastewaters based on the analytical procedure with flow process, being composed of a completely continuous flow system between a filtered sample solution and sulfuric acid (1+1) solution containing potassJum dichromate as oxidant. It also describes an discussion of the system by comparing with the FIA systems developed recently by the authors~l)-(s) on the continuous monitoring of COD in wastewaters having multiply, randomly, and small effluent

characteristics (e.g.,wastewater from university laboratories).

Experimental Reagents

Potassium dichromate (0.6 g), previously dried at about 105

°c

for 2 hr!4) is dissolved in 500 ml of distilled water and the solution is diluted with 500 ml of concentrated sulfuric acid. This solution is ca. 2 roM in dichromate and used as a reagent solution.

(3)

As a standard substance for COD, a mixture of L-glutamic acid and lactose (5:1) is used in this work. Various concentrations of aqueous standard solutions, for which CODM values have been determined by the

(10) n

official standard method, are prepared.

All chemicals used were of analytical-reagent grade or better.

Instrument and procedure

Figure 1 shows a schematic diagram of the system for continuous monitoring of COD. The system is constructed of commercially available parts for high performance liquid chromatography (HPLC) and

poly (tetrafluoroethylene) (PTFE) tubings, selected to withstand continuous operation for long periods of time.

A Kyowa-seimitsu, Model-KHU-W-52, double reciprocating micro-pump is used to pump both the dichromate solution and the filtrated sample solution at each flow rate of 0.3 ml/min. The sample is pre-treated with a 100-mesh stainless filter before pumping. The streams merge at a Kyowa-seimitsu, Model KZU-l, PTFE mixing joint. Reaction proceeds in a 20-m coil made of PTFE tubing (1 mm i.d., 3 mm o.d.) which is heated at 120 °C, in a TOYo, Model OC-24S, oil bath containing silicone oil.

The reaction mixture then passes into a quartz tubular flow-through cell (10 mm light pass, 18 pl volume) placed in a Shimadzu, Model 100-01, spectrophotometer. The absorbance measured at 445 nm, is recorded continuously using a Nippon-denshi-kagaku, Model U-228, multirange recorder. A calibration curve is prepared form the peak heights vs. the CODMn values of aqueous standard solutions.

Excess . Sample

._.

Filtration unit

Reagent soln.

t-Hcro-pump

r----··--- ..

• Reaction coil.

• •

I •

• • Waste

~-~."'"

,

• •

, .

! ..'

Spectrophotometr ic

...-._---

Oil bath detector

Fig. 1 Experimental apparatus for continuous monitoring of COD with complete flow process

(4)

Results and Discussion

Instrumentation

The apparatus for continuous monitoring of COD in wastewaters used in this work, is developed by a complete flow process as follows:

The apparatus system must, of course, be durable and reliable for a long time of operation. The peristaltic pumps, which are widely used in FIA and completely continuous flow

analyses~6)-(8)

are unsuitable because the present system is operated at least 5 atm (measured by a Kyowa-seimitsu, KPG-50N, pressure gauge) by connecting a back~pressure

PTFE coil (0.25 mm Ld., 3 m long) just after the flow cell. A double reciprocating micro-pump for HPLC is therefore used to propel both the filtered sample solution and the acidic dichromate solution.

In order to check a baseline drift, the filtered sample solution is exchanged to distilled water. However, a contamination of the flow cell was not observed in the present flow analysis system for COD.

Choice of operating conditions

In FIA, tubing diameters and lengths and flow rates must be chosen carefully to minimize dispersion for long reaction times of COD

measurements. In the present flow process, however, sample dispersion need not be considered since a portion of sample is not injected into the water stream necessary in the FIA system for COD~1)-(5)

Nevertheless, reagent consumption and waste quantity containing

chromium (VI) are still important. So, the micro-pump could be operated at a total flow rate of about 0.6 ml/min when the rather viscous

sulfuric acid (1+1) solution and water sample are separately pumped with reproducible flow rates. Each flow rate of 0.3 ml/min was selected to maintain a constant 1:1 ratio pumping. At these flow rates, a 20-m reaction coil of seamless 1.0-mm i.d. (3.0 mm o.d.) PTFE tubing was was optimal for the reaction, giving a residence time of about 30 min and a reasonable mixing between sample and reagent solutions. The quantity of waste was below 1 liter a day, causing few problem of waste treatment.

The effect of reaction'temperature on the oxidation with dichromate was examined by using the present standard substance obtained from

L-glutamic acid and lactose. The results showed that a temperature of 120

°c

was suitable~ temperature below 110

°c

lowered the sensitivity, and that between 120 and 140

°c

gave constant values, but that above

150

°c

might burst the weakened PTFE tubing owing to a high back-pressure.

(5)

Determination of COD with dichromate

Figure 2 shows a typical output from the present flow system with acidic potassium dichromate when different concentrations of aqueous standard solutions were pumped-up and distilled water was alternatively injected at every 30 min in order to check a base-line drift. A plot of peak heights obtained for the standard solutions under the

recommended operating conditions vs. the COD

Mn values determined by the official method~lO) was linear over the COD

Mn range 0-20 mg/l.

The detection limit of the method is 0.5 mg/l COD

Mn because a sample solution is not diluted with carrier solutions in comparison with the FIA method with dichromate~4)

As also shown in Fig. 2, the reproducibility of the present COD determination is good, reflecting the automatic character of the

complete flow process. There is little opportunity for personal errors since the apparatus is composed of an automatic sampling, reaction, and detection. The precision of the method is also good~ the standard

deviation was 0.4 % in 10 determinations of 15 mg/l CODMn standard sample solutions.

injection Baseline

omg/l

5 mg/l

10 mg/l

15 mg/l

60 min

•

20 mg/l as COD

Mn

Fig. 2 Typical output from the present flow system and calibration with the standard mixed both L-glutamic acid and lactose Application to continuous monitoring

Under the recommended conditions, the developed flow system was satisfactorily operated for several days. However, the permanganate method reported previously(6)-(8) produced a few precipitate of

(6)

17:00 19:00

Time (o'clock) 15:00

13:00 11:00

20 . . . ... .... ... &.. ~

9:00

.-i 8

"'-

0>

l::

OJ ::J .-i_rrJ 12

:>

2:c

0·

0 16

u 4

manganese (IV) oxide in a long run in spite of an employment of a sulfuric acid-phosphoric acid mixture. It was therefore found that potassium dichromate was effective in the continuous monit.oring of COD with the flow process.

To apply the present method. to a continuous monitoring of COD in wastewater samples, the effect of possible interfering ions must be known. Chloride ion, which present at a level of 10 mg/l in river water, was tested. As the results, it was found that chloride ion did not interfere up to 100 mg/l, even in the absence of silver salts, when potassium dichromate in sulfuric acid (1+1) solution was used as the oxidant solution in the present flow system. Whereas the flow method with acidic permanganate had to be removed an interference caused by chloride ion, by adding silver nitrate in the flow system~6)

The wastewater samples would require filtration with a 100-meshed stainless filter, and this would require great care to avoid altering the composition of the oxidizable organic substances present in water samples.

In the continuous monitoring of COD substances in wastewaters, the system can be fully automated by pumping both sample and reagent solutions with the double reciprocating micro-pump for HPLC. Typical examples of the results obtained from the wastewater samples of the Center for Environmental Science and Technology, Okayama University, are given in Fig. 3; samples were continuously pumped-up without an addition of distilled water for a base-line determination.

o

Fig. 3 Examples on continuous monitoring of COD

(7)

Comparison with the FIA method

In the FIA analyzer, each discrete sample is timely injected into the carrier (e.g.iwater) or reagent solutions!l)-(5) The advantage of FIA-type batch-flow approach is that at least 20 samples are analyzed for an hour because of timely injections of the sample solutions. The disadvantage is that the apparatus is mechanically complex to some extent with moving parts such as automatic sample injector. Hence, The batch-flow analyzers (i.e., FIA) are suitable for the analysis of large numbers of samples and the apparatus is less complex mechanically and is easier to construct than the batch analyzers available

commercially. (9) However, analytical results obtained from the FIA analyzers tend to deviate from those obtained by the standard method!lO) because the reaction in FIA does not simulate loyally. that in the batch process.

The complete flow analyzer represented here differs from batch flow analyzers such as FIA in only one respect. The sample is

continuously pumped-up as a.stream into a continuous reagent stream without injection parts at a similar flow rate. Disappointingly, the proposed method with complete flow process seems unadequate for the analysis of large numbers of wastewater samples.

Evaluation of this method as a continuous monitorih~ system

For the wastewater samples having multiply, randomly, and small effluent characteristics such as university laboratories, the completely continuous flow analyzer has some advantages over the batch analyzer, the air-segmented batch-flow analyzer, and the FIA-type batch-flow analyzer as a continuous monitoring system. The advantages obtained in this work are as follows.

(1) The present system can perform completely continuous and automated monitoring of environmental waters such as wastewater.

(2) It can precisely determine both total amounts present and concentrations at a given time.

(3) The amounts of reagents required in the present analyzer are much lower than for conventional batch analyzers, because the flow rates used for the reagents are about 0.3 ml/min.

(4) The quantity of reagent used and waste after measurement is therefore reduced to approximately 1/100 of that required for conventional batrih analyzers for continuous measurements.

(5) The completely continuous flow analyzer described here has fewer

(8)

moving parts, therefore the instrumentation is simpler and less costly than conventional batch analyzers or batch-flow analyzers (e.g~, FIA).

The authors are greatly indebted to a financial support by the Nissan Science Foundation.

References

(1) T. Korenaga and H. Ikatsu, Bunseki Kagaku, 31, (1982) 135.

(2) T. Korena9~~ H',Ikatsu, K. Masago, T. Moriwake, and T. Takahashi, Preprints of the 32ndAnnual Meeting of the Japan Society for

Analytical Che~istry, (1983) p.902.

(3) T. Korenaga, T. Moriwake, and T. Takahashi, Memoirs Sch. Eng.

Okayama Univ., 19-1, (1984) 53.

(4) T. Korenaga and H. Ikatsu, Anal. Chim. Acta, 141, (1982) 301.

(5) T. Korenaga, K. Okada, T. Moriwake, and T. Takahashi, Abstracts of the 1984 International Chemical Congress of Pacific Basin Societies,

(1984) 01A16.

(6) M. Goto, T. Shiroeda, and D. Ishii, Buseki Kagaku, 30, (1981) 403.

(7) M. Goto, Trends Anal. Chern., 2, (1983) 92.

(8) T. Korenaga, Kagaku to Kogyo, (1985) 551.

(9) T. Korenaga, Sanyo Gijutsu Zasshi,

11,

(1982) 1.

(10) JIS K 0102 {lQ81).